AI-Assisted Package Design for Improved Warpage Control of Ultra-Thin Packages

Package design process has evolved from the early days, when any design was subjected to vigorous analysis using Design of Experiments (DOE) methodology to understand the effect of each parameter and material change, to the introduction of finite element analysis (FEA) to reduce the DOE runs or completely eliminate it in some cases. FEA has proved to be a valuable tool. With the easy accessibility to computing power and the recent advances in AI-algorithms, this paper presents the next evolution in package design – AI-assisted package design. This framework incorporates both physics-driven and data-driven approaches to develop accurate and efficient design solutions. A validated AI-assisted package design framework is presented here to improve the design of the distribution of metal lines and layers in the substrate with the objective of reducing warpage. The framework consists of three phases – assembling the dataset, building a surrogate model to link substrate design to warpage and finally using global optimization routines and the surrogate model to propose changes to the metal layer density distribution across the substrate so as to reduce warpage.

[1]  Mingji Wang,et al.  Substrate Trace Modeling for Package Warpage Simulation , 2016, 2016 IEEE 66th Electronic Components and Technology Conference (ECTC).

[2]  Nagarajan Raghavan,et al.  Comparison of experimental, analytical and simulation methods to estimate substrate material properties for warpage reliability analysis , 2018, Microelectron. Reliab..

[4]  Dirk P. Kroese,et al.  Handbook of Monte Carlo Methods , 2011 .

[5]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[6]  Heikki Haario,et al.  DRAM: Efficient adaptive MCMC , 2006, Stat. Comput..

[7]  Raúl Rojas,et al.  Neural Networks - A Systematic Introduction , 1996 .

[8]  Nagarajan Raghavan,et al.  Machine Learning Approach to Improve Accuracy of Warpage Simulations , 2019, 2019 IEEE 69th Electronic Components and Technology Conference (ECTC).

[9]  J. Rosenthal,et al.  Markov Chain Monte Carlo , 2018 .

[10]  Pham Luu Trung Duong,et al.  Neural Network Assisted Speed Up of High Fidelity Warpage Simulations towards Design for Reliability in Ultra-Thin Packages , 2019, 2019 IEEE CPMT Symposium Japan (ICSJ).

[11]  Nagarajan Raghavan,et al.  Inverse Design of Substrate from Warpage Surrogate Model Using Global Optimisation Algorithms in Ultra-Thin Packages , 2020, 2020 IEEE 70th Electronic Components and Technology Conference (ECTC).

[12]  Ciyan Wu,et al.  Warpage Prediction Methodology of Extremely Thin Package , 2017, 2017 IEEE 67th Electronic Components and Technology Conference (ECTC).

[13]  Nagarajan Raghavan,et al.  Learning the Stress-Strain Relationships of Ultra-Thin Package Materials using a Bayesian Approach , 2020, 2020 IEEE 70th Electronic Components and Technology Conference (ECTC).

[14]  B. Han,et al.  Numerical/Experimental Hybrid Approach to Predict Warpage of Thin Advanced Substrates , 2018, 2018 IEEE 68th Electronic Components and Technology Conference (ECTC).

[15]  Methodology for Modeling Substrate Warpage Using Copper Trace Pattern Implementation , 2008, IEEE Transactions on Advanced Packaging.

[16]  Desmond J. Higham,et al.  Edinburgh Research Explorer Deep learning: an introduction for applied mathematicians , 2022 .

[17]  Seungbae Park,et al.  A Study of Substrate Models and Its Effect On Package Warpage Prediction , 2019, 2019 IEEE 69th Electronic Components and Technology Conference (ECTC).